Several years ago were developed transgenic mice in which the albumin (AL) enhancer/promoter was used to target expression of urokinase-type plasminogen activator (uPA) to hepatocytes. There were two dominant phenotypes in these mice: first, neonatal hemorrhaging in one-half of all transgenic progeny; second, uPA-induced hepatotoxicity in remaining progeny followed by overgrowth of the entire liver by clones of hepatocytes that had deleted transgene DNA and therefore no longer produced uPA. AL-uPA transgenic mice have become a unique model for the study of hepatic regeneration and neoplasia and have proven to be effective recipients for transplanted hepatocytes, permitting replacement of up to 80: of the liver parenchyma by donor mouse cells. More recently, immunocompromised AL-uPA mice have been used as recipients for transplanted rat liver cells, giving rise to mice bearing 100: rat hepatic parenchyma. These results form the starting point for the following proposal. The overall goal of this project is to develop a mouse model that does not display neonatal hemorrhaging in which mouse hepatocytes can be completely replaced by human hepatocytes. This would permit detailed experimental examination of human hepatocarcinogenesis in an in vivo setting, an approach unprecedented in the study of human liver biology and disease. Specifically, the aims are to: (1) generate and characterize transgenic mice with livers composed of human hepatocyte; (2) determine the hepatocarcinogenicity of genotoxic chemicals in humanized mouse liver; and (3) determine the hepatocarcinogenicity of non-genotoxic rodent liver carcinogens in humanized mouse liver. The results of this analysis should establish whether humanized mouse liver represents a more direct and meaningful model system in which to assess the human risk of potentially toxic chemical agents.